Transformation of Sign Language to Text in Digital Era Using Deep Neural Network

Sign language is a vital means of communication for the deaf and hard-of-hearing community, yet persistent communication barriers exist due to a lack of effective sign language interpretation tools. To address this, our study proposes “Sign2Text,” a Deep learning-based approach that recognizes and translates sign language into text in real-time, utilizing deep learning (DL) and computer vision methods with convolutional neural networks (CNNs) and SVM. ASL Dataset is considered in this research. The proposed architecture consists of hand segmentation and tracking, feature extraction using CNNs- MobileNetV1, sequence modelling and classification using SVM, Our model achieved an accuracy of 99.5% in recognizing individual signs and correctly predicted sign language sentences into text, thereby enhancing communication accessibility for the deaf and hard-ofhearing community.